Skip to contentSkip to conversation

tc reliably hangs my system

LARTC Dimitris Zilaskos 3 participants 8 messages Jul 3, 2002 – Jul 5, 2002

lore
Share & tools
lore
#0 lore raw

kernel 2.4.18-ac3 smp , iproute2-2.4.7-now-ss020116-try.tar.gz

I am using the following scripts to adjust my traffic during certain
periods of time , running them cron and sometimes by hand . They can reliably freeze the system if they are run certain number of times . No oops/panic/errors of whatsoever , just a frozen system .

the following "fast" script freezes the system after it is executed 7-8 times

tc qdisc del dev eth0 root
tc qdisc add dev eth0 root handle 10: cbq bandwidth 100Mbit avpkt 1000 tc class add dev eth0 parent 10:0 classid 10:1 cbq bandwidth 100Mbit \
rate 100Mbit allot 1514 weight 10Mbit prio 8 maxburst 20 avpkt 1000 tc class add dev eth0 parent 10:1 classid 10:100 cbq bandwidth 100Mbit \
rate 100Mbit allot 1514 weight 10Mbit prio 5 maxburst 20 avpkt 1000 tc class add dev eth0 parent 10:1 classid 10:200 cbq bandwidth 100Mbit \
rate 100Mbit allot 1514 weight 10Mbit prio 5 maxburst 20 avpkt 1000

tc qdisc add dev eth0 parent 10:100 sfq quantum 1514b perturb 10 tc qdisc add dev eth0 parent 10:200 tbf rate 6Mbit latency 50ms burst \
50000
tc qdisc del dev eth0 ingress
tc qdisc add dev eth0 handle ffff: ingress
tc filter add dev eth0 parent ffff: protocol ip prio 2 u32 match ip \
protocol 1 0xFF police rate 1kbit burst 10000 mtu 1k drop flowid :1

tc filter add dev eth0 parent 10: protocol ip prio 1 u32 match ip dst \
1.2.3.0/16 flowid 10:100
tc filter add dev eth0 parent 10: protocol ip prio 1 u32 match tcp src \
22 0xffff match ip protocol 0x6 0xff flowid 10:100
tc filter add dev eth0 parent 10: protocol ip prio 1 u32 match ip src \
1.2.3.25/32 flowid 10:200

The following "slow" script freezes the system after being executed
around 11-18 times

tc qdisc del dev eth0 root
tc qdisc add dev eth0 root handle 10: cbq bandwidth 100Mbit avpkt 1000 tc class add dev eth0 parent 10:0 classid 10:1 cbq bandwidth 100Mbit \
rate 100Mbit allot 1514 weight 10Mbit prio 8 maxburst 20 avpkt 1000 tc class add dev eth0 parent 10:1 classid 10:100 cbq bandwidth 100Mbit \
rate 100Mbit allot 1514 weight 10Mbit prio 5 maxburst 20 avpkt 1000 tc class add dev eth0 parent 10:1 classid 10:200 cbq bandwidth 100Mbit \
rate 100Mbit allot 1514 weight 10Mbit prio 5 maxburst 20 avpkt 1000

tc qdisc add dev eth0 parent 10:100 sfq quantum 1514b perturb 10 tc qdisc add dev eth0 parent 10:200 tbf rate 3Mbit latency 50ms burst \
50000
tc qdisc del dev eth0 ingress
tc qdisc add dev eth0 handle ffff: ingress
tc filter add dev eth0 parent ffff: protocol ip prio 2 u32 match ip \
protocol 1 0xFF police rate 1kbit burst 10000 mtu 1k drop flowid :1

tc filter add dev eth0 parent 10: protocol ip prio 1 u32 match ip dst \
1.2.3.0/16 flowid 10:100
tc filter add dev eth0 parent 10: protocol ip prio 1 u32 match tcp src \
22 0xffff match ip protocol 0x6 0xff flowid 10:100

tc filter add dev eth0 parent 10: protocol ip prio 1 u32 match ip src \
1.2.3.25/32 flowid 10:200

The ingress related parts are the prime suspects for me , i added them
recently . Before that no freezing has ever ocurred . I couldn't perform more tests because the box is a rather busy server and the lockups cause painfull reboots .

Any advice is welcomed .

Kind regards ,
--
======================================
Dimitris Zilaskos

Department of Physics @ Aristotle Univercity of Thessaloniki , Greece
======================================
_______________________________________________
LARTC mailing list / [email protected]
http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/

#1 lore raw
Quoted 7 lines · Dimitris Zilaskos

kernel 2.4.18-ac3 smp , iproute2-2.4.7-now-ss020116-try.tar.gz

I am using the following scripts to adjust my traffic during certain
periods of time , running them cron and sometimes by hand . They can reliably freeze the system if they are run certain number of times . No oops/panic/errors of whatsoever , just a frozen system .

Quoted 6 lines

The ingress related parts are the prime suspects for me , i added them
recently . Before that no freezing has ever ocurred . I couldn't perform more tests because the box is a rather busy server and the lockups cause painfull reboots .

Any advice is welcomed .

I ran both scripts for 10 minutes (while true; do ./test.sh; sleep 5; done) and no freeze for me.

Stef

Signature · 7 lines
[email protected]
 "Using Linux as bandwidth manager"
     http://www.docum.org/
     #lartc @ irc.openprojects.net
_______________________________________________
LARTC mailing list / [email protected]
http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/
#2 lore raw
In reply to Dimitris Zilaskos
Quoted 4 lines · Stef Coene

I ran both scripts for 10 minutes (while true; do ./test.sh; sleep 5; done) and no freeze for me.

Stef

I woke up early today to do some more tests . The hang occurs if there is
substantial traffic to the box running the scripts , ie around 300 kb/sec sustained outgoing traffic . The box serves as ftp server . I could not freeze it while the ftp daemon was not running (pureftpd) . When I started the ftp daemon and traffic began to flow , again running the scripts 7-8 times resulted in a frozen system . I am trying to find out the specific command that causes the problem....

Regards ,

--
======================================
Dimitris Zilaskos

Department of Physics @ Aristotle Univercity of Thessaloniki , Greece
======================================

_______________________________________________
LARTC mailing list / [email protected]
http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/

#3 lore raw
In reply to Dimitris Zilaskos

ok , this 2 lines repated anything from 5 to 20 times cause the hang :

tc qdisc del dev eth0 ingress
tc qdisc add dev eth0 handle ffff: ingress

again , the presence of sustained outgoing traffic catalyses the effect .
It takes at least 150-200 kbytes/sec to easily cause the hang .

Regards ,

--
======================================
Dimitris Zilaskos

Department of Physics @ Aristotle Univercity of Thessaloniki , Greece
======================================
_______________________________________________
LARTC mailing list / [email protected]
http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/

#4 lore raw
In reply to Dimitris Zilaskos
Quoted 8 lines · Dimitris Zilaskos

ok , this 2 lines repated anything from 5 to 20 times cause the hang :

tc qdisc del dev eth0 ingress
tc qdisc add dev eth0 handle ffff: ingress

again , the presence of sustained outgoing traffic catalyses the effect .
It takes at least 150-200 kbytes/sec to easily cause the hang .

I've forwarded this to the right kernel people - your description is specific enough, so I have good hope that this bug will be spotted soon!

Regards,

bert

Signature · 6 lines
http://www.PowerDNS.com          Versatile DNS Software & Services
http://www.tk                              the dot in .tk
http://lartc.org           Linux Advanced Routing & Traffic Control HOWTO
_______________________________________________
LARTC mailing list / [email protected]
http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/
#5 lore raw
In reply to Dimitris Zilaskos
Quoted 2 lines · bert hubert

I've forwarded this to the right kernel people - your description is specific enough, so I have good hope that this bug will be spotted soon!

Ok, I've been told that this might or might not help:

http://www.cyberus.ca/~hadi/patches/ing-stats.patch

This fixes a known issue that looks like this - can you try if this resolves your problem? This patch will be in in 2.4.19 probably.

Regards,

bert

Signature · 6 lines
http://www.PowerDNS.com          Versatile DNS Software & Services
http://www.tk                              the dot in .tk
http://lartc.org           Linux Advanced Routing & Traffic Control HOWTO
_______________________________________________
LARTC mailing list / [email protected]
http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/
#6 lore raw
In reply to Dimitris Zilaskos
Quoted 6 lines

http://www.cyberus.ca/~hadi/patches/ing-stats.patch

This fixes a known issue that looks like this - can you try if this resolves your problem? This patch will be in in 2.4.19 probably.

Thnx . I am using the patch now with success . I can no longer reproduce
the hang .

Regards ,

--
======================================
Dimitris Zilaskos

Department of Physics @ Aristotle Univercity of Thessaloniki , Greece
======================================
_______________________________________________
LARTC mailing list / [email protected]
http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/

#7 lore raw
In reply to Dimitris Zilaskos
Quoted 9 lines · Dimitris Zilaskos
Quoted 6 lines

http://www.cyberus.ca/~hadi/patches/ing-stats.patch

This fixes a known issue that looks like this - can you try if this resolves your problem? This patch will be in in 2.4.19 probably.

Thnx . I am using the patch now with success . I can no longer reproduce
the hang .

Good to hear - I reported the success to the author, Jamal Hadi Salim - thanks for verifying that it helps!

Regards,

bert

Signature · 6 lines
http://www.PowerDNS.com          Versatile DNS Software & Services
http://www.tk                              the dot in .tk
http://lartc.org           Linux Advanced Routing & Traffic Control HOWTO
_______________________________________________
LARTC mailing list / [email protected]
http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/